> ## Documentation Index
> Fetch the complete documentation index at: https://docs.sequenzy.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Start Live Sequence Test

> Test a saved sequence with trigger event properties, including cart IDs and arrays

Run your saved sequence for one active subscriber with an email address. This performs real actions, including webhooks, subscriber updates and discounts. Emails are marked as tests. It does not enable the sequence or record a trigger event. Requires `sequences:activate` and `subscribers:read`.

Pass `subscriberId` and optionally `speedMultiplier` (integer from 60 to 10000, default 120). Existing live-test delay and wait caps apply. Supply `customVariables` as the event properties object, without an outer `event` wrapper:

```json theme={null}
{
  "subscriberId": "sub_123",
  "customVariables": {
    "cart_id": "cart_123",
    "event_ids": ["event_1", "event_2"],
    "cart": { "total": 49.5 }
  }
}
```

Your sequence can reference `{{event.cart_id}}` and nested values such as `{{event.cart.total}}`. Arrays remain arrays. Omit `customVariables` or pass `{}` for no event properties; `null`, scalar roots and array roots are invalid. Supplying properties does not itself update subscriber attributes. These values belong to this run and are not saved as sequence configuration. The run response has no separate event-properties field, but resolved values can appear in step logs.

The response contains `run`, with `id`, `automationId`, `companyId`, `subscriberId`, `initiatedByUserId`, `recipientEmails`, `speedMultiplier`, `status`, `steps`, `errorMessage`, `createdAt`, `updatedAt`, `startedAt`, `finishedAt` and a creation-only `jobId`. Initial status is `queued`, steps are empty and start/finish/error fields are null. Poll [Get Live Sequence Test](./test-run-get) using `run.id`.

Invalid input returns 400; missing authentication returns 401; missing scopes or restricted roles return 403. Inaccessible sequences or active subscribers return 404. An existing queued/running test for the same sequence and subscriber returns 409. Queue failure returns 500 and marks the run failed.

Ordinary failure retries are disabled. Stalled-job recovery can replay actions after worker loss. A failed run can have completed real actions, which are not rolled back. Inspect its steps before starting a new run; repeating a run can repeat emails, webhooks and other actions.

```bash theme={null}
sequenzy sequences test-run seq_123 --subscriber sub_123 --event-properties '{"cart_id":"cart_123","event_ids":["event_1","event_2"]}' --json
sequenzy sequences test-run-get seq_123 run_123 --json
```

MCP provides `start_sequence_test_run` with the same request fields and `get_sequence_test_run` with `sequenceId` and `runId`. Use [Send Sequence Step Test](./send-test) for a single email or [Simulate Sequence](./simulate) for a dry run.